"No experiment can be more interesting than that we are now trying, and which we trust will end in establishing the fact, that man may be governed by reason and truth. Our first object should therefore be, to leave open to him all the avenues to truth. The most effectual hitherto found, is the freedom of the press. It is, therefore, the first shut up by those who fear the investigation of their actions." --Thomas Jefferson to John Tyler, 1804.

A Sali-Grant shift?

The new Greg Smith & Associates poll, just out this afternoon, has one startling result in it, one certain to start some debate rolling among Idaho political watchers.

It is so startling because little else in it is.

The last Smith poll, using the same methodology, emerged in early July. In the 1st District U.S. House race, the premier contest at the monent, that poll gave Republican Bill Sali a lead of 41% to 25% over Democrat Larry Grant, with the remainder (34%) undecided. The new poll, conducted shortly before Labor Day, gives Grant a lead – 22% to 14%. And 61% undecided.

That’s a double-headed stunner. If the numbers are right, then the number of undecideds in the race has nearly doubled, with almost all of that increase coming out of Sali’s hide. It doesn’t reflect acceleration on Grant’s part; his numbers are almost identical, even down by a hair. Nevertheless, if a mass of former Sali backers have started easing away, that ought to be a cause for deep concern on the part of that campaign. (Could it be related to the recent defensiveness we’ve noted earlier?)

Are the numbers right? Well, you’re never wise to put too much weight on any one single poll result. But consider this: All the other races polled both months – governor and lieutenant governor – show results almost unchanged from six weeks ago. And in two other races just polled, 2nd District U.S. House and superintendent of public instruction, the results show strong Republican leads, as you’d expect. Smith suggested that “The change is clearly a result of changing voter sentiment, not a change in polling methodology.”

We’d like to see whether this result is backed up a second time, or whether the next Smith poll swings back somewhat in the 1st CD. But the moment, with caveats noted, the poll suggests the 1st is a district to watch – one with a substantial number of voters not nailed down.

In the other results, there’s less of note.

Office Republican % Democrat %
Govenor Butch Otter 42% Jerry Brady 18%
Lt Gov Jim Risch 46% Larry La Rocco 23%
1st US House Bill Sali 14% Larry Grant 22%
2nd US House Mike Simpson 61% Jim Hansen 19%
Supt Publ Instr Tom Luna 40% Jana Jones 29%

No great surprises elsewhere, other than that the number of undecideds is holding fairly strong elsewhere, too. This may be an unsettled, and unsettling election cycle yet.

UPDATE: The Sali campaign apparently doesn’t handle bad news very well. It shot out a release contending, “The methodology used was so bad that Channel Seven, who commissioned the poll in the first place, has decided not to go with its obviously faulty results. We do not intend to release the results of our own internal polling at this time, for reasons of our own; but suffice it to say that our polling, using credible methodology and done by one of the most experienced and respected polling firms in the nation, bears no resemblance to Greg Smith’s conclusions.”

First, the Sali campaign had no issue with Smith’s numbers in July, which resulted from the same methodology (and showed Sali in a more favorable light). Second, the easiest refutation of Smith’s results would be release of the campaign’s internal polls, which the campaign says it has conducted, but which it is not releasing. Campaigns often release at least the base numbers – if, that is, releasing them puts the campaign in a favorable light. Campaigns that sit on poll numbers typically do so because they don’t look so hot – an argument in favor of Smith’s numbers.

The emotional response suggests that Smith’s results may have touched a nerve.

Share on Facebook


  1. Adam's Blog said:

    Wanted: A Poll with a Real Sample

    Ridenbaugh has the latest on a Greg Smith poll that has Larry Grant up over Bill Sali by 22-14% with Undecided at 64%.

    I’m skeptical of this poll and I was skeptical of the last one. The last poll had a margin of error of +/- 6%. This poll regardin…

    September 8, 2006
  2. adamelijah said:

    I’m very skeptical. I thought the first poll of 400 voters statewide and 218 of those in the first district was dubious at best. Now we get a poll with a margin of error of +/- 7.4% and only 300 voters statewide and probably half or a little more in the first district. It’s a tiny sample. It’s odd to have a five day poll produce 300 results. The last weekend of August right before Labor Day would not rank as a great time to do a poll either.

    I also wonder how Smith concludes as he does that Mike Simpson’s support is holding steady as he wasn’t in the July poll.

    Regardless, I look forward to not only seeing what the result of the next poll is, but hopefully we’ll get a poll that will have a larger sample. The tiny sample size that Smith has obtained so far have skewed any statistical significance. The first poll was better with a margin of +/- 6% but still incredibly limited in usefulness.

    September 8, 2006

Comments are closed.